Method and device for predicting a switching state and/or a switching time of a signaling system for traffic control

Abstract

A method predicts a switch state and/or a switch time point of a signaling system. The method includes collecting first and second state data, the first and second state data influencing the switch state and/or switch time point. The collection of the first state data includes reading of state data from a signaling system control device of the signaling system by a signaling system interface. The collection of the second state data includes reading in of the state data. A prediction model is provided and configured to make a prediction of the switch time point and/or the switch state of the signaling system based on first and second state data. The switch state and/or the switch time point of the signaling system is predicted via the prediction model using the first and second state data. The predicted switch state and/or switch time point of the signaling system is outputted.

The invention claimed is: 
     
       1. A method for predicting a switching state and/or a switching time of a signaling system for traffic control, which comprises the following method steps of:
 acquiring first state data and second state data, wherein the first state data and second state data influence the switching state and/or the switching time of the signaling system, wherein acquiring the first state data includes reading out state data of a signaling system controller of the signaling system by means of a signaling system interface, wherein acquiring the second state data includes reading in state data provided via a communication interface; 
 providing a prediction model via a central computing unit via a cloud interface, the prediction model configured to make a prediction of the switching time and/or the switching state of the signaling system depending on the first state data and second state data; 
 predicting the switching state and/or the switching time of the signaling system by means of the prediction model on a basis of the first state data and second state data; 
 outputting a predicted switching state and/or switching time of the signaling system; 
 receiving a termination command via the cloud interface and the outputting of the predicted switching state and/or switching time of the signaling system is ended after the termination command has been received; and 
 outputting a confidence interval for the predicted switching state and/or switching time. 
 
     
     
       2. The method according to  claim 1 , wherein the first state data includes control data of the signaling system. 
     
     
       3. The method according to  claim 1 , which further comprises:
 determining a control recommendation for the signaling system on a basis of the predicted switching state and/or switching time; and 
 outputting the control recommendation to the signaling system via the signaling system interface. 
 
     
     
       4. The method according to  claim 1 , which further comprises carrying out remote maintenance of the signaling system controller via the signaling system interface. 
     
     
       5. The method according to  claim 1 , wherein a second state data read in via the communication interface includes data from external sensors and/or data from internet-based third-party providers and/or data from roadside units for obtaining vehicle data. 
     
     
       6. The method according to  claim 1 , which further comprises outputting the predicted switching state and/or switching time to vehicles. 
     
     
       7. The method according to  claim 1 , which further comprises repeating the acquiring of the first state data and/or of the second state data after a predefined time duration, wherein the predefined time duration is a maximum of five seconds. 
     
     
       8. The method according to  claim 1 , which further comprises outputting the first state data and/or the second state data to the central computing unit via the cloud interface. 
     
     
       9. The method according to  claim 1 , wherein after the termination command has been received, a resumption command is received via the cloud interface, and after the resumption command has been received, the outputting of the predicted switching state and/or switching time of the signaling system is resumed. 
     
     
       10. The method according to  claim 1 , which further comprises training the prediction model by means of the first state data and/or the second state data. 
     
     
       11. The method according to  claim 1 , wherein the prediction model is data-driven. 
     
     
       12. The method according to  claim 1 , which further comprises controlling a vehicle on a basis of the predicted switching state and/or switching time. 
     
     
       13. A device, comprising:
 a signaling system interface; 
 a communication interface; and 
 a computing unit configured to carry out the method according to  claim 1 . 
 
     
     
       14. A non-transitory computer-readable storage medium having computer executable instructions for performing the method according to  claim 1 . 
     
     
       15. A system, comprising:
 said device according to  claim 13 ; and 
 a vehicle configured to receive the predicted switching state and/or switching time, said vehicle having a vehicle controller configured to control a vehicle movement of said vehicle on a basis of the switching state and/or switching time. 
 
     
     
       16. A method for evaluating a prediction model, which comprises the following steps of:
 providing a previous prediction model, the previous prediction model being output to a device via a computing unit interface; 
 reading first state data and/or second state data into a central computing unit via the computing unit interface; 
 comparing a switching state and/or a switching time calculated by means of the previous prediction model with an actual switching state and/or an actual switching time; and 
 outputting a termination command via the computing unit interface if a calculated switching state and/or a calculated switching time deviates from the actual switching state and/or the actual switching time to a given extent. 
 
     
     
       17. The method according to  claim 16 , which further comprises subsequently training the prediction model on a basis of the first state data and/or the second state data, wherein after the prediction model has been trained, a newly trained prediction model is stored in a memory. 
     
     
       18. The method according to  claim 17 , which further comprises:
 performing a check to ascertain whether the newly trained prediction model is better suited to predicting the switching state and/or the switching time than the previous prediction model; 
 outputting the newly trained prediction model and a resumption command via the computing unit interface if the newly trained prediction model is better suited to predicting the switching state and/or the switching time than the previous prediction model; and 
 outputting the resumption command via the computing unit interface if the newly trained prediction model is worse suited to predicting the switching state and/or the switching time than the previous prediction model. 
 
     
     
       19. A central computing unit configured to carry out the method according to  claim 16 . 
     
     
       20. A non-transitory computer-readable storage medium having computer executable instructions for performing the method according to  claim 16 .